Case Study: Auditing the Mission Design of the Mars Reconnaissance Orbiter

DataField.Dev

Case Study: Auditing the Mission Design of the Mars Reconnaissance Orbiter

"Aerobraking is the difference between a rocket you can afford and one you cannot."

Executive Summary

The Mars Reconnaissance Orbiter (MRO) has circled Mars since 2006, returning more data than every other Mars mission combined and serving as the communications relay for the rovers on the surface. It is a textbook mission — literally, for this chapter — because its architecture is the completed Track-C MDR of §40.4 made real: a Hohmann cruise, a single make-or-break orbit-insertion burn, and then a six-month aerobraking campaign that traded propellant for time. In this case study we audit MRO the way a review board would: we reconstruct its delta-v budget, check that the vehicle closes on its launch vehicle, and quantify the single design decision — aerobraking — that made the mission launchable at all. We are not designing anything; we are checking someone else's finished design and asking why it is the shape it is.

Skills applied

Reconstructing an interplanetary delta-v budget from the Hohmann transfer (§40.4, Ch. 11).
Computing a Mars orbit-insertion burn from the arrival hyperbola (§40.4, Ch. 6).
Auditing an aerobraking-vs-propulsive trade and its mass consequence (Ch. 29).
Checking closure against a real launch vehicle (Ch. 30, Appendix H).

A note on numbers. Every MRO figure below is Tier 2 — widely reported, approximate, and not pinned to one primary source. Masses, burn sizes, and dates vary by a few percent between sources. We audit the architecture and its logic, which are robust, not the third significant figure, which is not.

Background

MRO's objective was to study Mars from a low, near-polar orbit at unprecedented resolution — its HiRISE camera resolves objects the size of a desk from orbit — and to relay data for surface missions. That objective flows down, exactly as Chapter 29 describes, to a specific orbit (a ~$255 \times 320\ \text{km}$ near-polar science orbit), which sets a delta-v budget, which sizes a vehicle, which picks a launch vehicle. MRO launched on an Atlas V 401 from Cape Canaveral in August 2005, cruised ~7 months, and reached Mars in March 2006 with a launch mass of roughly $2{,}180\ \text{kg}$, of which about $1{,}150\ \text{kg}$ was propellant (hydrazine). Our audit asks: does that architecture hang together, and where is its leverage?

Phase 1: Reconstruct the interplanetary leg

MRO flew a near-minimum-energy Hohmann transfer, the very trajectory computed in Chapter 11 and reused in §40.4. The launch vehicle (Atlas V plus its Centaur upper stage) supplied the departure energy — a characteristic energy of roughly $C_3 \approx 9\text{–}17\ \text{km}^2/\text{s}^2$ depending on the year's geometry — placing MRO on a trans-Mars trajectory. The spacecraft's own delta-v does not begin until Mars.

The key arrival number is the hyperbolic excess speed at Mars, $v_\infty \approx 2.9\ \text{km/s}$ (Tier 2, close to the $2.65\ \text{km/s}$ of the idealized §40.4 transfer; the real value depends on the launch date). This is the speed MRO had relative to Mars as it fell in — the speed it had to shed to be captured.

Phase 2: Compute the orbit-insertion burn

At the periapsis of its arrival pass (~$300\ \text{km}$ altitude, radius $r \approx 3{,}690\ \text{km}$, Mars $\mu = 4.283\times10^{4}\ \text{km}^3/\text{s}^2$), MRO's speed on the hyperbola was

$$ v_{\text{hyp}} = \sqrt{v_\infty^2 + \frac{2\mu}{r}} = \sqrt{2.9^2 + \frac{2(4.283\times10^{4})}{3{,}690}} = \sqrt{8.41 + 23.21} = 5.62\ \text{km/s}. $$

Two possible capture targets bracket the decision:

Just-capture into a highly elliptical orbit (periapsis speed = escape speed $\sqrt{2\mu/r} = 4.82\ \text{km/s}$): burn $\Delta v = 5.62 - 4.82 = 0.80\ \text{km/s}$ (a little more, ~$1.0\ \text{km/s}$, to reach a useful, more-bound capture ellipse).
Propulsive capture straight to the low circular science orbit (circular speed $\sqrt{\mu/r} = 3.41\ \text{km/s}$): burn $\Delta v = 5.62 - 3.41 = 2.22\ \text{km/s}$.

MRO's actual MOI burn was about $\mathbf{1.0\ \text{km/s}}$ — a ~27-minute burn on its main engines into a highly elliptical capture orbit (about $426 \times 44{,}500\ \text{km}$). It captured, but nowhere near its final orbit. Everything from that ellipse down to the science orbit was done without propellant.

🔗 Connection: This is precisely the capture trade of §40.4, and MRO chose the cheap branch. Capturing to the ellipse cost ~$1.0\ \text{km/s}$; going straight to the low circular orbit would have cost ~$2.2\ \text{km/s}$ — more than twice as much. The difference, ~$1.2\ \text{km/s}$, is what aerobraking bought back.

Phase 3: Audit the aerobraking decision

Over roughly five to six months and about 445 passes, MRO dipped its periapsis into the thin Martian upper atmosphere on each orbit, letting drag shave a little velocity off apoapsis each time, walking the far point of its orbit down from $44{,}500\ \text{km}$ to a few hundred. This is the ~$1.2\ \text{km/s}$ of orbit-lowering that would otherwise have been propulsive — supplied instead by aerodynamics, for the price of time and care.

Quantify the mass consequence with the rocket equation. Suppose MRO's dry mass is ~$1{,}030\ \text{kg}$ (launch mass minus propellant) and its hydrazine system has $I_{sp} \approx 230\ \text{s}$ ($v_e = 2{,}256\ \text{m/s}$). The propellant to supply an extra $1.2\ \text{km/s}$ of capture burn, on top of everything else, would have multiplied the wet mass by a factor $e^{1200/2256} = e^{0.53} = 1.70$ — that is, a 70% larger vehicle just for the capture leg it skipped.

import math
G0 = 9.80665

def wet_mass(dv, isp, dry):
    """Wet mass (kg) to supply delta-v dv (m/s) at specific impulse isp (s) for dry mass (kg)."""
    return math.exp(dv / (isp * G0)) * dry

DRY = 1030.0            # kg, MRO dry mass (Tier 2)
ISP = 230.0             # s, hydrazine monopropellant (Tier 2)

# Extra propellant multiplier for the ~1.2 km/s aerobraking replaced:
factor = math.exp(1200 / (ISP * G0))
print(f"extra wet-mass factor for +1.2 km/s capture: {factor:.2f}x")

# Illustrative: a same-dry vehicle sized for propulsive-only capture (~2.2 km/s + ~0.3 corrections)
prop_only = wet_mass(2500, ISP, DRY)
aerobrake = wet_mass(1300, ISP, DRY)     # ~1.0 capture + ~0.3 corrections
print(f"propulsive-only wet: {prop_only:.0f} kg")
print(f"aerobraking wet:     {aerobrake:.0f} kg")
print(f"aerobraking saved:   {prop_only - aerobrake:.0f} kg")
# Expected output:
# extra wet-mass factor for +1.2 km/s capture: 1.70x
# propulsive-only wet: 3143 kg
# aerobraking wet:     1785 kg
# aerobraking saved:   1358 kg

The audit's central finding: aerobraking saved on the order of 1.3 tonnes of vehicle mass (Tier 2, illustrative). That is the difference between a spacecraft an Atlas V 401 can send to Mars and one that would need a substantially larger, costlier launch vehicle — or a smaller science payload. One trade decision, one box in the funnel, reshaped the whole mission.

Phase 4: Check closure on the launch vehicle

Does the design close? An Atlas V 401 can send roughly $2{,}300\text{–}2{,}600\ \text{kg}$ toward Mars at the required $C_3$ (Tier 2, Appendix H). MRO's ~$2{,}180\ \text{kg}$ launch mass fits, with margin — but only because it aerobraked. The ~$3{,}100\ \text{kg}$ propulsive-only vehicle from Phase 3 would have overrun the 401's capability, forcing a bigger, more expensive Atlas V configuration (more solid boosters) or a redesign. The design closes on the cheaper launcher precisely because of the aerobraking decision. This is the closure check of §40.1 done on a real vehicle: sized wet mass under launch capacity, with margin.

🔧 Engineering Reality: Aerobraking is not free even setting aside the six months. Each pass heats the spacecraft and loads its solar arrays aerodynamically; fly too deep and you overheat or tear a panel, too shallow and you make no progress and waste months. MRO's team adjusted periapsis altitude daily, pass by pass, threading a corridor only a few kilometres wide in an atmosphere whose density varies with Martian weather. The propellant saving is real, but it is bought with operational intensity and genuine risk — exactly the "propellant vs. time and risk" trade §40.4 named.

Phase 5: Identify the driving requirement

Trace the sensitivities and the driving requirement emerges: MRO's science orbit — low, near-polar, and circular, for the high-resolution imaging that was the mission's reason to exist — is what forced a large velocity change at Mars, which forced either a heavy propulsive capture or a long aerobraking campaign. The instrument (HiRISE and its need for a low orbit) drove the architecture. Change that one requirement — accept a higher, elliptical science orbit — and the capture burn shrinks, aerobraking becomes unnecessary, and the whole vehicle changes. The imaging resolution requirement is the input the coupled design was most sensitive to, and everything else followed from serving it.

⚠️ Common Misconception: "MRO aerobraked because aerobraking is better." Aerobraking is better for a mission that needs a low circular orbit, can afford six months, and can accept the operational risk. A mission that needed to begin science immediately, or could not risk the atmospheric passes, would rationally choose propulsive capture and a bigger launch vehicle. The audit's lesson is the same as every trade study in the book: the right answer is the output of the driving requirement and the weighting, not a universal rule.

Discussion Questions

MRO captured into a highly elliptical orbit rather than going straight to its science orbit. Explain, in terms of the arrival hyperbola of Phase 2, why the elliptical capture is so much cheaper in delta-v.
The audit estimated aerobraking saved ~1.3 tonnes. Which single quantity in the wet_mass calculation drives that saving, and through which equation?
Suppose a future orbiter must reach its science orbit within days of arrival (say, to observe a transient event). How would that requirement change the capture architecture, and what would it cost?
MRO used hydrazine at $I_{sp} \approx 230\ \text{s}$. If it had instead used a bipropellant engine at $I_{sp} \approx 320\ \text{s}$, roughly how would the capture propellant change? (Hand-trace the mass ratio for the ~1.0 km/s capture burn at both.)

Your Turn: Extensions

Option A (analysis). Recompute the Phase 2 insertion burn for a higher arrival altitude — say a $500\ \text{km}$ periapsis instead of $300\ \text{km}$. Does capturing higher up cost more or less delta-v, and why? (Recompute $v_{\text{hyp}}$, $v_{\text{esc}}$, and $v_c$ at the new radius; add # Expected output:.)
Option B (computation). Extend the Phase 3 script to sweep the aerobraking-replaced delta-v from 0 to 1.5 km/s and print the wet mass at each step, reproducing the mass-vs-aerobraking curve. Where does aerobraking stop being worth the operational trouble?
Option C (your mission). If your track is C, audit your Mars orbiter against MRO: is your capture propulsive or aerobraked, and does your sized vehicle close on your chosen launch vehicle? If another track, identify the analogous "propellant-vs-time" trade in your design and audit whether you took the cheaper branch.

Key Takeaways

A real mission is the MDR spine made physical. MRO is objective → orbit → delta-v budget → vehicle → launch vehicle, exactly as §40.1 lays it out, with aerobraking as the pivotal trade.
The capture burn is set by the arrival hyperbola. $v_{\text{hyp}} = \sqrt{v_\infty^2 + 2\mu/r}$, minus the target orbit's speed; capturing to an ellipse is far cheaper than to a low circular orbit.
Aerobraking trades propellant for time and risk. It supplied ~1.2 km/s of orbit-lowering aerodynamically, saving on the order of a tonne of vehicle mass and letting MRO fly on a cheaper Atlas V — the closure decision of the whole mission.
Architectures serve the driving requirement. The low-orbit imaging requirement drove the large capture, hence the aerobraking, hence the launch-vehicle choice. Change that one requirement and the whole vehicle changes.